The constant demand for seafood products and the undeniable effects of fishing on marine ecosystems make it urgent to implement an ecosystem approach, even in data-poor scenarios such as small-scale fisheries. Understanding the impacts of fishing is essential for promoting management strategies that prevent irreversible damage to marine ecosystems. Thus, ecosystem quantitative science-based models have been frequently used to evaluate the effects of fishing, although fishers’ local ecological knowledge (LEK) can aid the implementation of qualitative models, particularly in data-poor conditions. Here, we present a framework for simulating and assessing the effects of fishing following two strategies: (1) for both types of models, we simulated species removal scenarios, and (2) for quantitative science-based models, we fitted time series to dynamically assessed impacts. The impacts were analyzed through ecological indicators commonly used for quantitative models, and because these indicators cannot be easily estimated for qualitative models, we propose the use of topological indicators in both types of models. The approach was applied to three case studies of small-scale finfish fisheries in northwestern Mexico. We found that the ecosystem response to species removal was different in each case study and that the target species can play an important role in ecosystems, but their removal does not generate abrupt changes in the ecosystem structure. The quantitative science-based models were able to reproduce the historical catch trends, which allowed us to reveal that changes in ecosystems are indeed influenced by fishing effort but also by underlying primary productivity. Furthermore, topological and ecological indicators showed similar trends in the quantitative models, which suggests that the former could be useful when data-poor conditions allow only qualitative models. This result confirms the relevance of the participation of fishers in generating qualitative models and their decisive role in the discussion of co-management strategies and risk scenarios in a better-informed manner.